The short answer
A ring modulator does one strange, simple thing: it multiplies your voice by a second tone called the carrier. It does not filter your voice, and it does not just lay another sound on top of it. Multiplication is a different operation from mixing, and it produces something no equalizer or delay could make: the hard, metallic, robotic character you know from decades of science-fiction voices.
The reason is a piece of math. When you multiply a tone at one frequency by a carrier at another, neither original frequency survives. In their place you get two new tones: one at the difference of the two frequencies and one at their sum. These new tones are called sidebands, and a pure ring modulator throws the original voice away and keeps only them. Because the sidebands usually land on frequencies with no simple relationship to your voice, your ear hears them as inharmonic, which is the clangorous, bell-like sound the effect is known for.
Multiplication, not mixing
It is worth being precise about that word. When you mix two sounds, you add them: play a note and a carrier together and you hear both, side by side, each keeping its own pitch. A ring modulator does not add, it multiplies: at every instant it takes the value of your voice and scales it by the value of the carrier. Multiplying two waves that swing above and below zero is a much wilder operation than adding them, and the frequencies that come out are not the frequencies that went in.
There is a clean rule for what comes out. Multiply a tone at frequency A by a carrier at frequency C and you get exactly two tones: one at A minus C and one at A plus C. Nothing else. The identity below is the whole engine of the effect.
sin(A) * sin(C) = 0.5 * [ cos(A - C) - cos(A + C) ]Multiply a tone by a carrier and you get a difference tone (A minus C) and a sum tone (A plus C); the originals are gone.
Put numbers on it and it stops being abstract. Suppose a piece of your voice sits at 300 Hz and you set the carrier to 200 Hz. Out come two tones, one at 100 Hz (the difference) and one at 500 Hz (the sum), while the original 300 Hz vanishes. Every tone in your voice gets this same treatment at once, each one split into a pair and shifted, which is why the whole sound changes character rather than just changing pitch.
The two sidebands
A picture of the frequencies makes the idea concrete. Think of a spectrum as a row of spikes, one for each tone present, standing taller where a tone is louder and placed left to right by pitch. Feed one tone into a ring modulator and watch what the spectrum does.
The spacing is the carrier, so the carrier frequency is the single control that sets the whole sound. Nudge it and both sidebands slide outward or inward together, which is why turning one knob can walk the effect from a low grumble to a high, glassy chime.
This also explains why the result is inharmonic. Any pitched sound is a stack of harmonics: a fundamental plus overtones at whole-number multiples of it, and that neat ladder is exactly what your ear reads as a single, natural pitch. Ring modulation shifts every one of those harmonics by the same fixed amount, the carrier, rather than scaling them, so the tidy multiples are gone, replaced by tones with no common fundamental. Whole-number ratios read as musical; shifted, unrelated frequencies read as clangorous, the way a struck bell or a sheet of metal does. That is the ring modulator's signature.
Ring modulation versus tremolo
A ring modulator has a close cousin worth knowing, because the difference explains one of its most useful controls. If, instead of throwing the original away, you keep it and add the sidebands around it, you have amplitude modulation. And when the carrier is slow, amplitude modulation is just tremolo: a steady rise and fall in volume. Drop a ring modulator's carrier below about 30 Hz, down toward the speeds of a slow LFO, and you stop hearing two separate tones and start hearing a fast chop in loudness, drifting into that tremolo territory.
Two honest notes follow from this. First, patchd does not ship a dedicated tremolo node; the Ring Modulator at a very low carrier is the nearest thing, drifting toward amplitude-modulation territory, but it is not a tremolo. Second, this cousin relationship is the trick to staying intelligible: a pure ring mod removes your voice, so words can get lost inside the metallic tones, but bleeding some dry voice back in keeps your original pitch present underneath the sidebands, so the words stay readable while the robot character sits on top. One point of vocabulary while we are here: tremolo is periodic change of volume, while vibrato is periodic change of pitch. They are endlessly confused, and some vintage gear even swapped the labels, but they are opposite things.
The carrier frequency is the whole sound
Because everything hangs on the carrier, it is worth knowing roughly what each range does. These are not hard lines, but a reliable map.
- Under about 30 Hz. The two sidebands sit so close to the original that you hear a choppy wobble in volume rather than distinct tones, the tremolo-like edge described above.
- Around 50 to 500 Hz. The classic zone. The sidebands are far enough apart to sound clearly inharmonic, giving the hard, buzzy, robotic voice, a classic sci-fi robot voice, that most people picture when they think of a ring modulator.
- Higher still. The sidebands spread far apart and take on a bright, ringing, bell-and-metal quality, more clangorous chime than voice.
The carrier's shape matters too. A plain sine carrier is the textbook case that gives exactly the two sidebands above; richer carrier shapes carry their own harmonics into the multiplication and pile up many more sidebands, for a denser, harsher tone.
The ring modulator in patchd
In patchd, the Ring Modulator is one node in the Studio voice-fx rack, sitting in the same picker as the reverb, delay, and phaser nodes. You add it to a channel like any other DSP node and dial it in with a small set of controls that map straight onto the theory on this page.
- Frequency sets the carrier, and so it sets everything: low for a choppy wobble, the middle range for the classic robot, higher for a clangorous bell.
- Waveform chooses the carrier shape. Sine is the textbook two-sideband ring mod; other shapes add more sidebands for a denser, more aggressive sound.
- Mix sets how much dry voice bleeds back in. Raising it keeps your original pitch present under the metallic tones, morphing the effect toward amplitude modulation so words stay intelligible.
- Low-cut trims rumble off the modulated signal, since a low carrier can push energy down into a muddy, boomy range.
It is the same node you would reach for to build a robot or a machine voice, and it lives alongside the rest of the voice changer effects in a rack you can order however you like. The engine is real time and runs over the low-overhead Windows audio paths at single-digit milliseconds of engine-internal latency at a small ASIO buffer, so a Ring Modulator node never gets in the way of monitoring yourself live.
The takeaway
A ring modulator multiplies your voice by a carrier tone, and that single operation replaces each tone with a pair of sidebands, one at the sum and one at the difference of the two frequencies, while a pure ring mod removes the original. Because those sidebands rarely fall on whole-number multiples of your voice, the result is inharmonic: metallic, robotic, and bell-like. The carrier frequency is the whole sound, and bleeding some dry voice back keeps your words readable underneath it.
patchd is a pre-launch Windows audio mixer with a full effects rack on every channel and a Studio voice-fx rack that includes the Ring Modulator among its nodes. If you want a mixer that can turn your voice metallic and robotic in real time while the rest of your audio stays clean, join the waitlist and get notified when it ships. The Studio tier is $39.99 per year, and the real-time engine is identical on both tiers.